Download earthquakes and earth`s interior

CHAPTER 10
EARTHQUAKES AND EARTH’S INTERIOR
CHAPTER OUTLINE
What Is an Earthquake?
Earthquake Destruction
Seismology
Can Earthquakes Be Predicted?
Locating the Source of an
Earthquake
Probing Earth’s Interior
Measuring the Size of Earthquakes
Discovering Earth’s Major
Boundaries
Destruction Destruction caused by a severe earthquake that struck Haiti on January 12, 2010. About
100,000 people perished and 600,000 were left homeless.
(Photo by Thony Belizaire/AFP/Getty Images)
10_tarb_ch10.indd ccxxxiii
11/11/10 9:13 AM
234
Chapter 10
Earthquakes and Earth’s Interior
them, and shantytowns were swept away by landslides. The quake flattened hillsides, knocked out
communications, blocked roads, destroyed most
hospitals, and clogged Haiti’s main seaport with
debris. The disaster was made even worse by strong
aftershocks that rumbled for days after the main
event. “It was a worst-case scenario. It should be a
wake-up call for the entire Caribbean,” lamented
an earthquake expert. Haiti, Dominican Republic,
and Jamaica are sandwiched between strike-slip
faults (Figure 10.1C), like the famous San Andreas.
The quake occurred along a 65-kilometre stretch of
the southern fault (near Port-au-Prince), where the
Caribbean plate had been stuck against the smaller
Gonvave platelet for 240 years. On January 12 the
“They are pulling people out of the rubble, literally
with blood running in the gutter like water,” sighed
an exhausted aid worker in Haiti, the western hemisphere’s poorest country, which was devastated by a
magnitude 7.0 earthquake in January 2010 (chapter
opening photo; Figure 10.1). Some survivors huddled together; others lay in the street, unable to pick
themselves up. Many spent desolate hours wandering in shock while others cowered under makeshift
shelters as night fell.
Such was the scene late in the afternoon of
January 12, 2010, around the capital city of Port-auPrince (Figure 10.1A, B). It was difficult to take in
the sheer scale of destruction as poorly constructed
buildings crumpled like paper with people still in
A
B
Atlantic Ocean
North America Plate
Cuba
Septentrional Fault
Haiti
telet
Cayman Gonvave Pla
Jamaica
Trough
in
lo-Planta
Enriquil
lt
u
a
F
Garden
Caribbean Sea
Honduras
0
250
500 km
Dominican
Republic
Puerto
Rico
Epicentre of
January 12, 2010
Haiti Earthquake
Caribbean Plate
C
◆ Figure 10.1 A. Devastation in downtown Port-au-Prince after the January 12, 2010, earthquake. B. Poorly built houses
crumble on a hillside above Port-au-Prince; others were swept away by landslides. The image in C shows the geology around
Haiti, which is sandwiched between left-lateral strike-slip faults. The quake occurred near Port-au-Prince along the EnriquilloPlantain Garden fault that separates the Caribbean plate and Gonvave platelet. When the plates moved in opposite directions,
built-up energy was violently released.
(Photo A by REUTERS/Jorge Silva; photo B by REUTERS/Eduardo Munoz; C modified from an image courtesy of Science Daily)
10_tarb_ch10.indd 234
11/11/10 9:13 AM
What Is an Earthquake?
two plates jerked forward about 2 metres in opposite
directions to relieve built-up stress, thereby releasing
massive amounts of energy in Haiti. Because it was
a land-based quake, no major tsunami was generated like the famous one triggered by the Boxing
Day earthquake under the Indian Ocean in 2004. In
1946 a more powerful 8.1 quake struck the northeast corner of Haiti, but the damage was greater in
2010 because the quake was near the capital city, it
was shallow, and buildings were not constructed to
withstand shaking. This is also why Haiti’s 7.0 quake
was more devastating than Chile’s 8.8 quake the following month.
Earthquakes and associated phenomena, such
as landslides and tsunami, are tangible reminders
that Earth is a highly dynamic system, both above
and below its surface. Each year, more than a thousand earthquakes occur in Canada alone, although
the majority of these produce vibrations that are
simply too weak to be felt by humans and can be
detected only by sensitive instruments.
Occasionally, media reports alert us to particularly strong earthquake events. The most severe
earthquake disasters in recent times have included
those centred in Izmit, Turkey (August 17, 1999,
with more than 17,000 deaths); Gujarat, India
(January 26, 2001, with more than 20,000 deaths);
the Bam region of Iran (December 26, 2003, with
more than 25,000 deaths); and Sumatra, Indonesia
(December 26, 2004, with 230,000 deaths, many by
tsunami). Disasters like these are difficult for most
Canadians to comprehend; indeed, many Canadians
have never even felt an earthquake in their lifetime.
The events that occurred in the Indian Ocean and
Haiti are particularly significant in that they changed
the consciousness of millions of people worldwide.
Victims were reported in at least eight countries
bordering the Indian Ocean after the 2004 event.
In a sense, these disasters were truly felt by the
global community. Before the Boxing Day disaster,
few people had even heard the term tsunami; by the
next day, it had become a household word. It was a
strong reminder that earthquake hazards can extend
far beyond the area in which an earthquake itself is
generated. These events were also wake-up calls to
the many nations that, having not experienced an
earthquake for some time, had become complacent
about preparing for such a disaster.
Ancient stories of the Aboriginal peoples of
Vancouver Island, historical documents from Japan,
computer-modelling methods, and a variety of geologic evidence indicate that an earthquake of similar
magnitude to the Sumatra-Andaman earthquake
occurred in 1700. Canada is not immune to tsunami: a rare Atlantic tsunami associated with the
10_tarb_ch10.indd 235
235
1929 Grand Banks, Newfoundland and Labrador,
earthquake was responsible for at least 10 deaths—
historically, the largest number of deaths attributed
to a single Canadian earthquake. Thus, the question
is not “will another earthquake occur in Canada?”
but, rather, “when will the next big earthquake
happen—and where?” Answering this question will
require knowledge of past earthquake events and the
study of current earthquake activity.
In this chapter, we examine what earthquakes
are, how they are produced, the damage they cause,
and how they are studied. We also explore the positive side of earthquakes—how the behaviour of seismic waves has been used to deduce the structure of
Earth’s interior.
WHAT IS AN EARTHQUAKE?
GEODe:Earth
Earthquakes
What Is an Earthquake?
An earthquake is the vibration of Earth produced
by the rapid release of energy. Most often, earthquakes are caused by slippage along faults in Earth’s
crust. The energy released radiates in all directions
from its source—the focus (focus ⫽ a point)—in the
form of seismic waves. These waves are analogous to
those produced when a stone is dropped into a calm
pond (Figure 10.2). Just as the impact of the stone
sets water waves in motion, an earthquake generates
seismic waves that radiate throughout Earth. Even
though the energy dissipates rapidly with increasing distance from the focus, sensitive instruments
located around the world record the event.
Fault scarp
Epicentre
Wave fronts
Focus
Fault
◆ Figure 10.2 Earthquake focus and epicentre. The focus
is the place within Earth at which the initial displacement
occurs. The epicentre is the surface location directly above
the focus.
11/11/10 9:14 AM
236
Chapter 10
Earthquakes and Earth’s Interior
Earthquakes and Faults
The tremendous energy released by atomic explosions or by volcanic eruptions can produce earthquakes, but these events are relatively weak and
infrequent. What mechanism produces a destructive
earthquake? Ample evidence exists that Earth is not
a static planet. We know that Earth’s crust has been
uplifted at times, because we have found numerous
ancient wave-cut benches many metres above the
level of the highest tides. Other regions exhibit evidence of extensive subsidence. In addition to these
vertical displacements, offsets in fence lines, roads,
and other structures indicate that horizontal movement is common (Figure 10.3). These movements
are usually associated with faults (Chapter 9).
Most of the motion along faults can be satisfactorily explained by plate tectonic theory, as introduced in Chapter 1. Mobile plates of lithosphere
interact with neighbouring plates, straining and
deforming the rocks at their margins. In fact, it is
along the faults associated with plate boundaries that
most earthquakes occur. Furthermore, earthquakes
are repetitive: as soon as one is over, the continuous
motion of the plates resumes, adding stress to the
rocks until they fail again.
Elastic Rebound
The actual mechanism of earthquake generation
eluded geologists until H. F. Reid of Johns Hopkins
University conducted a study following the great
1906 San Francisco earthquake. The earthquake was
accompanied by horizontal surface displacements of
several metres along the northern portion of the San
Andreas Fault. This 1300-kilometre fracture extends
northwest–southeast through southern California.
It is a large fault zone that separates two great sections of Earth’s crust: the North American plate and
the Pacific plate. Field investigations determined
that during this single earthquake, the Pacific plate
lurched as much as 5 metres northward past the adjacent North American plate.
The mechanism for earthquake formation that
Reid deduced from this information is illustrated in
Figure 10.4. In part A of the figure, you see an existing fault. In part B, tectonic forces slowly deform the
crustal rocks on both sides of the fault, as demonstrated by the bent features. Under these conditions,
rocks bend and store elastic energy, much like a
wooden stick does if bent. Eventually, the frictional
resistance holding the rocks together is overcome. As
slippage occurs at the weakest point (the focus), displacement exerts stress farther along the fault, where
additional slippage occurs (Figure 10.4C) until most
of the built-up stress is released (Figure 10.4D). This
slippage allows the deformed rock to “snap back.”
The vibrations we know as an earthquake occur
as the rock elastically returns to its original shape.
The springing back of the rock was termed elastic
rebound by Reid because the rock behaves elastically, much like a stretched rubber band does when it
is released. After release, stress builds up again along
the fault as the earthquake cycle continues.
In summary, most earthquakes are produced by
the rapid release of elastic energy stored in rock that
has been subjected to great stress. Once the strength
of the rock is exceeded, it suddenly ruptures, causing
the vibrations of an earthquake. Earthquakes most
commonly occur along existing faults whenever the
frictional forces on the fault surfaces are overcome.
◆ Figure 10.3 This fence
was offset 2.5 metres
during the 1906 San
Francisco earthquake.
(Photo by G. K. Gilbert,
U.S. Geological Survey)
10_tarb_ch10.indd 236
11/11/10 9:14 AM
Seismology
Deformation of rocks
Stream
Fault
A Original position
237
they can sometimes destroy already badly weakened
structures. This occurred during a 1988 earthquake
in Armenia and the 2010 Haitian quake (chapter
opening photo; Figure 10.1). A large aftershock collapsed many structures that had been weakened by
the main tremor.
In addition, small earthquakes called foreshocks often precede a major earthquake by days
or, in some cases, by as much as several years.
Monitoring of these foreshocks has been used as a
means of predicting forthcoming major earthquakes,
with mixed success. We will consider the topic of
earthquake prediction later in this chapter.
SEISMOLOGY
Fault
GEODe:Earth
B Buildup of strain
C Slippage
D Strain released
◆ Figure 10.4 Elastic rebound. As rock is deformed, it
bends, storing elastic energy. Once strained beyond its
breaking point, the rock fails (breaks), releasing the stored-up
energy as earthquake waves.
Elastic Rebound
Foreshocks and Aftershocks
The intense vibrations of the 1906 San Francisco
earthquake lasted about 40 seconds. Although most
of the displacement along the fault occurred in this
rather short span, additional movements along this
and other nearby faults lasted for several days following the main quake. The adjustments that follow a
major earthquake often generate smaller earthquakes
called aftershocks. Although these aftershocks are
usually much weaker than the main earthquake,
10_tarb_ch10.indd 237
Earthquakes
Seismology
The study of earthquake waves, seismology
(seismos ⫽ shake, ology ⫽ the study of) dates back to
attempts made by the Chinese almost 2000 years ago
to determine the direction from which these waves
originated. The seismic instrument used by the
Chinese was a large hollow jar that contained a suspended mass (similar to a clock pendulum) connected
in some fashion to the jaws of several large dragon
figurines that encircled the container (Figure 10.5).
The jaws of each dragon held a metal ball. When
earthquake waves reached the instrument, the relative motion between the suspended mass and the jar
would dislodge some of the metal balls into the waiting mouths of frog figurines directly below.
The Chinese were probably aware that the
first strong ground motion from an earthquake is
directional, and when it is strong enough, all poorly
supported items will topple over in the same direction. Apparently the Chinese used this fact, plus the
position of the dislodged balls, to detect the direction to an earthquake’s source. However, the complex motion of seismic waves makes it unlikely that
the actual direction to an earthquake was determined
with any regularity.
In principle, modern seismographs (seismos ⫽
shake, graph ⫽ write), instruments that record seismic waves, are not unlike the device used by the
early Chinese. Seismographs have a mass freely suspended from a support that is attached to the ground
(Figure 10.6). When the vibration from a distant
earthquake reaches the instrument, the inertia*
*Inertia: Simply stated, objects at rest tend to stay at rest, and objects
in motion tend to remain in motion, unless acted on by an outside force.
You probably have experienced this phenomenon when an automobile
you were in stopped quickly and your body continued to move forward.
11/11/10 9:14 AM
238
Chapter 10
Earthquakes and Earth’s Interior
◆ Figure 10.5 Ancient Chinese seismograph. During an
Earth tremor, the dragons located in the direction of the main
vibrations would each drop a ball into the mouth of the frog
below it.
(Photo by James E. Patterson)
(iners ⫽ idle) of the mass keeps it relatively stationary
while Earth and the support move. The movement of
Earth in relation to the stationary mass is recorded on a
rotating drum, magnetic tape, or digital device.
The records obtained from seismographs,
called seismograms (seismos ⫽ shake, gramma ⫽
what is written), provide a great deal of information
concerning the behaviour of seismic waves. Seismic
waves are energy forms that radiate out in all directions from the focus. The propagation (transmission)
of this energy can be compared to the shaking of
gelatine in a bowl that results as some is spooned
out. Whereas the gelatine will have one mode of
vibration, seismograms reveal that two main groups
of seismic waves are generated by the slippage of
a rock mass. The wave types that travel through
Earth’s interior are called body waves. The types
that travel along the outer part of Earth are called
surface waves. Body waves are further divided into
two types, called primary waves, or P waves, and
secondary waves, or S waves.
Body waves are divided into P and S waves by
their mode of travel through intervening materials.
P waves are “push–pull” waves—they push (compress) and pull (expand) rocks in the direction the
wave is travelling (Figure 10.7A). Imagine holding
someone by the shoulders and shaking that person.
This push–pull movement is how P waves move
through Earth. Solids, liquids, and gases resist
◆ Figure 10.6 Principle of the seismograph. The inertia of the suspended
mass tends to keep it motionless, while the recording drum, which is anchored
to bedrock, vibrates in response to seismic waves. Thus the stationary mass
provides a reference point from which to measure the amount of displacement
occurring as the seismic wave passes through the ground. A. Seismograph
designed to record horizontal movement. B. Seismograph
designed to record vertical ground motion.
Seismographs
Weight hinged
to allow movement
A
B
Spring
Support
moves with
Earth
Bedrock
Mass does not
move with
ground motion
because of
inertia
Pen
Support
moves
Rotating
drum
Hinge
Suspended
mass
Anchored base
Rotating drum
records motion
Bedrock
Bedrock
Earth moves
10_tarb_ch10.indd 238
Vertical ground motion
11/11/10 9:14 AM
Seismology
a change in volume when compressed and will
elastically spring back once the force is removed.
Therefore, P waves, which are compressional waves,
can travel through all these materials.
239
Conversely, S waves “shake” the particles at
right angles to their direction of travel. This can be
illustrated by fastening one end of a rope and shaking the other end, as shown in Figure 10.7B. Unlike
Coil spring at rest
Push coil spring
Compress
Wave direction
Expand Compress
Wave direction
Particle motion
A P wave
Rope at rest
Shake rope
Wave direction
Particle motion
Wave direction
Particle motion
B S wave
C Surface wave
D Surface wave
◆ Figure 10.7 Basic types of seismic waves and their characteristic motion. (Note that during a typical earthquake, ground
shaking consists of a combination of various kinds of seismic waves that have complex movements in three dimensions.) A. As
illustrated by a coil spring, P waves are compressional waves that alternately compress and expand the material through which
they pass. The back-and-forth motion produced as compressional waves travel along the surface can cause the ground to buckle
and fracture, and they can cause power lines to break. B. S waves cause material to oscillate at right angles to the direction of
wave motion. Because S waves can travel in any plane, they produce up-and-down and sideways shaking of the ground. C. One
type of surface wave is essentially the same as that of an S wave that exhibits only horizontal motion. This kind of surface wave
moves the ground from side to side and can be particularly damaging to the foundations of buildings. D. Another type of surface
wave travels along Earth’s surface, much like rolling ocean waves. The arrows show the elliptical
Seismic Wave Motion
movement of rock as the wave passes.
10_tarb_ch10.indd 239
11/11/10 9:14 AM
240
Chapter 10
Earthquakes and Earth’s Interior
◆ Figure 10.8 Typical
seismogram. Note the time
interval (about 5 minutes)
between the arrival of the first
P wave and the arrival of the
first S wave.
Surface waves
First P wave
First S wave
1 minute
10_tarb_ch10.indd 240
The first travel-time graphs were greatly improved
when seismograms became available from nuclear
explosions, because the precise location and time of
detonation were known.
2000
1000
16
3000 m
15
14
13
12
First S wave
cu
rv
e
11
10
5 min
av
e
9
w
P waves, which temporarily change the volume of
intervening material by alternately compressing and
expanding it, S waves temporarily change the shape
of the material that transmits them. Because fluids
(gases and liquids) do not respond elastically to
changes in shape, they will not transmit S waves.
The motion of surface waves is somewhat
more complex (Figure 10.7C). As surface waves
travel along the ground, they cause the ground and
anything resting on it to move, much as ocean swells
toss a ship. In addition to their up-and-down motion,
surface waves have a side-to-side motion similar to
an S wave oriented in a horizontal plane. This latter
motion is particularly damaging to the foundations
of structures.
By observing a “typical” seismic record, as
shown in Figure 10.8, you can see major differences
among these seismic waves: P waves arrive at the
recording station first, then S waves, and then surface waves. This is a consequence of their speeds.
To illustrate, the velocity of P waves through granite
within the crust is about 6 kilometres per second.
S waves under the same conditions travel at 3.6 kilometres per second. Differences in density and elastic
properties of the rock greatly influence the velocities of these waves. Generally, in any solid material,
P waves travel 1.7 times as fast as S waves, and
S waves travel 1.1 times as fast as surface waves.
A system was developed for locating earthquake epicentres by using seismograms from earthquakes whose epicentres could be easily pinpointed
from physical evidence. From these seismograms,
travel-time graphs were constructed (Figure 10.9).
(Later)
8
S-
T I M E
Travel time (minutes)
(Earlier)
First P wave
ve
ur
ec
v
wa
P-
7
6
5
4
3
2
1
0
1000
2000
3000
4000 km
Distance to epicentre
◆ Figure 10.9 A travel-time graph is used to determine the
distance to the epicentre. The difference in arrival times of
the first P and S waves in the example is 5 minutes. Thus, the
epicentre is roughly 3800 kilometres away.
11/11/10 9:14 AM
Locating the Source of an Earthquake
In addition to velocity differences, also notice
in Figure 10.8 that the height or, more correctly, the
amplitude of these wave types varies. The S waves
tend to have a slightly greater amplitude than the
P waves, while the surface waves, which cause the
greatest destruction, tend to exhibit an even greater
amplitude. Because surface waves are confined to a
narrow region near the surface and are not spread
throughout Earth as P and S waves are, they retain
their maximum amplitude longer. Surface waves also
have longer periods (time intervals between crests);
therefore, they are commonly referred to as long
waves, or L waves.
As you will see, seismic waves are useful in
determining the location and magnitude of earthquakes, in addition to providing information on the
nature of Earth’s internal characteristics.
LOCATING THE SOURCE OF
AN EARTHQUAKE
GEODe:Earth
Earthquakes
Locating the Source of an Earthquake
Recall that the focus is the place within Earth
where earthquake waves originate. The epicentre
(epi ⫽ upon, centr ⫽ a point) is the location on the
surface directly above the focus (see Figure 10.2).
The difference in velocities of P and S waves
provides a method for locating the epicentre. The
principle used is analogous to a race between two
Paris
Montréal
241
vehicles, one faster than the other. The P wave
always wins the race, arriving ahead of the S wave.
But the greater the length of the race, the greater
will be the difference in the arrival times at the finish line (the seismic station). Therefore, the greater
the interval measured on a seismogram between the
arrival of the first P wave and the first S wave, the
greater the distance to the earthquake source. By
using the sample seismogram in Figure 10.8 and
the travel-time curves in Figure 10.9, we can determine the distance separating the recording station
from the earthquake in two steps: (1) determine
the time interval between the arrival of the first
P wave and the first S wave, and (2) find on the
travel-time graph the equivalent time spread between
the P and S wave curves. From this information,
we can determine that this earthquake occurred
3800 kilometres from the recording instrument.
Now we know the distance but what direction?
The epicentre could be in any direction from the
seismic station. As shown in Figure 10.10, the precise
location can be found when the distance is known
from three or more different seismic stations. On a
globe, we draw a circle around each seismic station.
Each circle represents the epicentre distance for each
station. The point at which the three circles intersect
is the epicentre of the quake. This method is called
triangulation.
The study of earthquakes was greatly bolstered
during the 1960s through efforts to discriminate
between underground nuclear explosions and natural
84
km
670
0k
00
m
◆ Figure 10.10 An earthquake
epicentre is located by using the
distances obtained from three or
more seismic stations.
Epicentre
m
0k
550
São Paulo
10_tarb_ch10.indd 241
11/11/10 9:14 AM
242
Chapter 10
Earthquakes and Earth’s Interior
earthquakes. A worldwide network of more than
100 seismic stations is coordinated through Golden,
Colorado. The largest of these, located in Billings,
Montana, consists of an array of 525 instruments
grouped in 21 clusters covering a region 200 kilometres in diameter. By using data from this array
and high-speed computers, seismologists are able
to differentiate between nuclear blasts and natural
earthquakes, as well as determine the position of an
earthquake’s epicentre.
Earthquake Belts
About 95 percent of the energy released by earthquakes originates in a few relatively narrow zones
that wind around the globe (Figure 10.11). The
greatest energy is released along a path around the
outer edge of the Pacific Ocean known as the circumPacific belt. Included in this zone are regions of great
seismic activity, such as Japan, the Philippines, Chile,
and numerous volcanic island chains, as exemplified
by the Aleutian Islands.
Another major concentration of strong seismic
activity extends through the mountainous regions that
flank the Mediterranean Sea and continues through
Iran and on past the Himalayan complex. Figure 10.11
indicates that yet another continuous belt extends for
thousands of kilometres through the world’s oceans.
This zone coincides with the oceanic ridge system, an
area of frequent but weaker seismic activity.
Areas of North America included in the
circum-Pacific belt lie adjacent to California’s San
Andreas Fault and along the western coastal regions
extending to the Aleutian Islands. In addition to these
high-risk areas, other sections of North America are
regarded as regions where strong earthquakes are
likely to occur. In Canada earthquakes are concentrated in British Columbia but do occur elsewhere
(Box 10.1).
Earthquake Depths
Areas of North America included in the circumPacific belt lie adjacent to California’s San Andreas
Fault and along the western coastal regions extending
to the Aleutian Islands. In addition to these high-risk
areas, other sections of North America are regarded as
regions where strong earthquakes are likely to occur.
In Canada earthquakes are concentrated in British
Columbia but do occur elsewhere (Box 10.1).
Evidence from seismic records reveals that
earthquakes originate at depths ranging from 5
to nearly 700 kilometres. In a somewhat arbitrary
fashion, earthquake foci have been classified according to their depth of occurrence. Those with points
of origin within 70 kilometres of the surface are
referred to as shallow, while those generated between
70 and 300 kilometres are considered intermediate,
and those with a focus deeper than 300 kilometres
are classified as deep. About 90 percent of all earthquakes occur at depths of less than 100 kilometres,
and nearly all very damaging earthquakes appear to
originate at shallow depths (the 2010 Haitian one
originated 13 kilometres deep).
◆ Figure 10.11 Distribution of the 14,229 earthquakes with magnitudes of at
least 5 between 1980 and 1990.
(Data from National Geophysical Data Center/NOAA)
10_tarb_ch10.indd 242
Plate Boundary Features
11/11/10 9:14 AM
Locating the Source of an Earthquake
BOX 10.1
243
C A N A D IA N PROFIL E
Earthquakes in Canada
David Eaton*
On August 22, 1949, residents of
the Queen Charlotte Islands were
jolted by Canada’s largest measured
earthquake. With a magnitude of 8.1,
this offshore tremor ranks as one of
the great earthquakes of the twentieth century, well recorded around
the globe. The earthquake knocked
cows off their feet and caused an oil
tank at Cumshewa Inlet to collapse.
Even 300 kilometres away in Terrace,
British Columbia, the ground shaking
was so intense that standing on the
street was described as “like being
on the heaving deck of a ship at sea.”
Fortunately, the sparse population
of the region meant that casualties
and property damage were light in
comparison with other earthquakes
of similar magnitude, such as the
1946 Vancouver Island earthquake
(Figure 10.12B).
The 1949 earthquake ruptured a
major segment of the Queen Charlotte fault system, an active rightlateral transform fault along Canada’s
west coast (Figure 10.A). Across the
fault, the Pacific plate is creeping
slowly northward relative to North
America, in a manner analogous to
lithospheric movement along the San
Andreas Fault system in California.
Friction across the fault resists this
motion, causing stress to accumulate
during aseismic intervals. The sudden release of stored energy during
an earthquake (elastic rebound) starts
the earthquake cycle once again.
In southwest British Columbia, a
different plate-tectonic setting exists,
more akin to Japan than California
(Figure 10.A). From southern Vancouver Island to northern California, the
oceanic Juan de Fuca plate is subducting beneath the North American
plate within a system known as the
Cascadia subduction zone. Although
large earthquakes have taken place,
such as the widely felt Nisqually
*David Eaton is a professor in the Department
of Earth Sciences at the University of Calgary.
10_tarb_ch10.indd 243
North American
plate
Pacific
plate
Magnitude
8–9
7–8
6–7
5–6
Juan de
Fuca plate
Partial
melting
Oceanic
lithosphere
Continental
lithosphere
Source of the great
earthquake of A.D. 1700
Sub
duc
tion
Partial
melting
◆ Figure 10.A Configuration of lithospheric plates and locations of significant earthquakes
that have been recorded in British Columbia. (Modified from R. J. W. Turner, J. J. Clague, and
B. J. Groulx, Geoscape Vancouver, Living with Our Geological Landscape III, Geological Survey
of Canada, 1998, Open File 3309; reproduced with the permission of the Minister of Public
Works and Government Services Canada, 2008, and courtesy of Natural Resources Canada)
◆ Figure 10.B The distribution and magnitudes of earthquakes in Canada since 1980.
(Reproduced with the permission of Natural Resources Canada, 2010, and courtesy of
Earthquake Canada)
earthquake near Seattle, Washington,
on February 28, 2001, the Cascadia
subduction zone is unusual because
no large historical earthquake has
occurred along the dipping plate
boundary. (The Nisqually earthquake
took place within the oceanic slab as
it broke apart during descent into
the mantle). There is now compelling
scientific evidence, however, that a
11/11/10 9:14 AM
244
Chapter 10
Earthquakes and Earth’s Interior
so-called megathrust earthquake
took place with an estimated magnitude of 9.0 along the plate boundary on January 26, 1700, before
settlement of the region by Europeans. The timing of this earthquake
is known from the oral traditions of
the Cowichan people of Vancouver
Island, from the study of tree rings
in remnant stumps of drowned forests in estuaries along the Oregon
and Washington coast, and from
meticulous Japanese written records
of a devastating tsunami. By using
computer simulations, scientists have
demonstrated that the tsunami must
have originated near the west coast
of North America.
The giant Cascadia earthquake of
1700 was not a unique event. Studies
of shoreline sediments from Oregon
to Vancouver Island reveal a cyclic
pattern of sudden downdrop of the
coast, followed by a slow rebound
in elevation. In the last 6000 years,
13 such great earthquakes have been
documented. We are presently in
the midst of a long span of episodic
stress buildup between megathrust
earthquake events, producing coastal
uplift and a squeezing of Vancouver
Island that is measurable by using
modern survey techniques. The next
offshore megathrust earthquake will
happen, perhaps hundreds of years in
the future, and residents of southwest
British Columbia and adjacent areas
need to be ready for it.
◆ Figure 10.C On June 23, 2010, tremors from a magnitude 5.0 earthquake centred near
Val-des-Bois, Quebec (approximately 60 km north of Ottawa), were felt over much of eastern
and southern Ontario and northeastern United States. Minor damage to buildings and roads
were experienced within tens of kilometres of the epicentre. This bridge, located a few
kilometres south of Val-des-Bois, collapsed as a direct result of the earthquake.
(Photo by Jean Levac, The Ottawa Citizen)
The tectonically active west coast
is not the only part of Canada that
experiences earthquakes. Figure 10.B
shows a map of earthquakes in Canada
since 1980, catalogued as part of the
National Earthquake Hazards Program
of the Geological Survey of Canada. In
western Canada, zones of earthquake
activity are found in the Rocky Mountains of southwest Alberta and the
Mackenzie Mountains of the Northwest
Territories, where earthquakes of magnitudes up to 6.6 have been recorded.
These bands of seismicity are thought
to be the result of transmission of plate
boundary stresses from convergent
margins (Cascadia and Alaska), far into
the North American plate.
Although very far from any
active plate boundary, parts of eastern Canada also experience mod-
MEASURING THE SIZE OF
EARTHQUAKES
Historically, seismologists have employed a variety
of methods to obtain two fundamentally different
measures that describe the size of an earthquake:
intensity and magnitude. The first to be used was
intensity—a measure of the degree of earthquake
shaking at a given locale based on the amount of
damage. With the development of seismographs,
it was possible to obtain a quantitative measure of
an earthquake based on seismic records rather than
uncertain personal estimates of damage. The measurement that was developed, called magnitude,
relies on calculations that use data provided by
seismic records (and other techniques) to estimate
10_tarb_ch10.indd 244
erate earthquakes, such as the
magnitude 5.9 Saguenay earthquake
on November 25, 1988, and more the
recent magnitude 5.0 Val-des-Bois
earthquake on June 23, 2010 (Figure
10.C). The seismicity of eastern Canada is mainly concentrated along the
Ottawa and St. Lawrence River valleys,
and in the Appalachians of Quebec
and New Brunswick. Like intraplate
earthquakes in the United States,
including the New Madrid earthquake
sequence in 1811–1812, the cause of
these earthquakes is not well understood. The zones of seismicity appear
to be associated with the locations
of ancient rifted margins or failed rift
zones, which may rupture (reactivate)
in response to the slow movement of
North America away from the spreading mid-Atlantic ridge.
the amount of energy released at the source of the
earthquake.
As it turns out, both intensity and magnitude
provide useful, although quite different, information about earthquake strength. Consequently, both
measures are still used to describe the relative sizes
of earthquakes.
Intensity Scales
Until a little more than a century ago, historical
records provided the only accounts of the severity
of earthquake shaking and destruction. The use
of these descriptions—which were compiled without any established standards for reporting—made
accurate comparisons of earthquake sizes difficult,
11/11/10 9:14 AM
Measuring The Size of Earthquakes
TABLE 10.1
245
Modified Mercalli Intensity Scale
I
Not felt except by a very few under especially favourable circumstances.
II
Felt by a few persons at rest, especially on upper floors of buildings.
III
Felt quite noticeably indoors, especially on upper floors of buildings.
IV
During the day, felt indoors by many, outdoors by few. Sensation like heavy truck striking building.
V
Felt by most people, many awakened. Disturbances of trees, poles, and other tall objects sometimes noticed.
VI
Felt by all; many frightened and run outdoors. Some heavy furniture moved; few instances of fallen plaster or
damaged chimneys. Damage slight.
VII
Everybody runs outdoors. Damage negligible in buildings of good design and construction; slight to moderate in
well-built ordinary structures; considerable in poorly built or badly designed structures.
VIII
Damage slight in specially designed structures; considerable in ordinary substantial buildings with partial collapse;
great in poorly built structures (fall of chimneys, factory stacks, columns, monuments, walls).
IX
Damage considerable in specially designed structures. Buildings shifted off foundations. Ground cracked conspicuously.
X
Some well-built wooden structures destroyed. Most masonry and frame structures destroyed. Ground badly cracked.
XI
Few, if any (masonry) structures remain standing. Bridges destroyed. Broad fissures in ground.
XII
Waves seen on ground surfaces. Objects thrown upward into air. Damage total.
(Table 10.1) was developed by using California
buildings as its standard, but it is appropriate for
use throughout most of the United States and
Canada to estimate the strength of an earthquake
(see Figure 10.12). For example, if some well-built
wooden structures and most masonry buildings
are destroyed by an earthquake, a region would
be assigned an intensity of X on the Mercalli scale
(Table 10.1).
at best. To standardize the study of earthquake
severity, early earthquake investigators developed
various intensity scales that considered damage
done to buildings, as well as individual descriptions
of the event, and secondary effects—landslides and
the extent of ground rupture. By 1902, Giuseppe
Mercalli had developed a relatively reliable intensity scale, which in a modified form is still used
today. The Modified Mercalli Intensity Scale
BRITISH
COLUMBIA
Prince Rupert
ALBERTA
Campbell River
IV
Chimneys collapse
Courtenay
Chimneys collapse
Post office wall
collapses
Epicentre
Port Alberni
Chimneys collapse
Water mains break
Power out for 10 days
V
VII
VI
B
Vancouver
Lions Gate Bridge sways
Seattle
Buildings sway
Portland
0
300
Kilometres
◆ Figure 10.12 A. Modified Mercalli Intensity map showing
intensity of shaking associated with an earthquake (magnitude
7.3) that occurred at 10:15 A.M. on June 23, 1946. The
epicentre was in central Vancouver Island, near the communities
of Courtenay and Campbell River. B. Damage to a school
in Courtenay, British Columbia, resulting from the 1946
earthquake.
(Part A © Department of Natural Resources Canada. All rights
reserved. Photo B reproduced with the permission of Natural
Resources Canada, 2010)
A
10_tarb_ch10.indd 245
11/11/10 9:15 AM
246
Chapter 10
Earthquakes and Earth’s Interior
Despite their usefulness in providing seismologists with a tool to compare earthquake severity, particularly in regions without seismographs,
intensity scales have severe drawbacks. In particular,
intensity scales are based on effects of earthquakes
(largely destruction) that depend not only on the
severity of ground shaking but also on external factors, such as population density, building design, and
the nature of surface materials. In British Columbia,
for example, the 1946 Vancouver Island earthquake
caused more property damage (Figure 10.12B) than
the much stronger earthquake that occurred west
of the Queen Charlotte Islands in 1949, largely
because it affected a more densely populated area.
Haiti (2010) experienced intensity of VII-IX on the
Mercalli scale.
Magnitude Scales
To compare earthquakes across the globe, a measure was needed that does not rely on parameters
that vary considerably from one part of the world to
another, such as types of construction. As a consequence, the Richter scale was developed.
RICHTER MAGNITUDE In 1935, Charles Richter of
the California Institute of Technology developed
Amplitude
23 mm
24 s
P
S
Distance, S–P,
km
s
500
50
400
40
0
10
200
100
60
40
20
0.5
10_tarb_ch10.indd 246
20
20
Magnitude,
ML
Amplitude,
mm
100
6
50
5
20
300
30
30 mm
10
Seismograph record
Time
s
the first magnitude scale by using seismic records
to estimate the relative sizes of earthquakes. As
shown in Figure 10.13, the Richter scale is based
on the amplitude of the largest seismic wave (P, S,
or surface wave) recorded on a seismogram. Because
seismic waves weaken as the distance between the
earthquake epicentre and the seismograph increases
(in a manner similar to light), Richter developed
a method that accounted for the decrease in wave
amplitude with increased distance. Theoretically,
as long as the same, or equivalent, instruments
were used, monitoring stations at various locations
would obtain the same Richter magnitude for every
recorded earthquake (Richter selected the WoodAnderson seismograph as the standard recording
device). In practice, however, different recording
stations often obtained slightly different Richter
magnitudes for the same earthquake—a consequence
of the variations in rock types through which the
waves travelled.
Although the Richter scale has no upper limit,
the largest magnitude recorded on a Wood-Anderson
seismograph was 8.9. These great shocks released an
amount of energy that is roughly equivalent to the
detonation of 1 billion tonnes of TNT. Conversely,
earthquakes with a Richter magnitude of less than 2.0
are not felt by humans. Earthquakes vary enormously
10
20
4
5
10
8
6
3
2
4
2
0.5
1
1.2
1
2
0.1
0
◆ Figure 10.13 Illustration
showing how the Richter
magnitude of an earthquake
can be determined graphically
by using a seismograph record
from a Wood-Anderson
instrument. First, measure
the height (amplitude) of the
largest wave on the seismogram
(23 millimetres) and then the
distance to the focus by using
the time interval between S
and P waves (24 seconds).
Next, draw a line between the
distance scale (left) and the wave
amplitude scale (right). By doing
this, you obtain the Richter
magnitude (ML) of 5.
(Data from California Institute of
Technology)
11/11/10 9:15 AM
Measuring the Size of Earthquakes
TABLE 10.2
Earthquake
Magnitude
247
Earthquake Magnitude and Energy Equivalence
Energy Released
(millions of ergs)
0
1
2
3
4
5
6
630,000
20,000,000
630,000,000
20,000,000,000
630,000,000,000
20,000,000,000,000
630,000,000,000,000
7
8
20,000,000,000,000,000
630,000,000,000,000,000
9
20,000,000,000,000,000,000
10
630,000,000,000,000,000,000
Approximate Energy Equivalence
0.5 kilograms of explosives
Energy of lightning bolt
450 kilograms of explosives
1946 Bikini atomic bomb test
1994 Northridge earthquake
1989 Loma Prieta earthquake
1906 San Francisco earthquake
1980 Eruption of Mount St. Helens
1964 Alaskan earthquake
1960 Chilean earthquake
Annual U.S. energy consumption
Source: U.S. Geological Survey.
Note: For each unit increase in magnitude, the energy released increases 31.6 times.
in strength, and great earthquakes produce wave
amplitudes that are thousands of times larger than
those generated by weak tremors. To accommodate
this wide variation, Richter used a logarithmic scale to
express magnitude, where a tenfold increase in wave
amplitude corresponds to an increase of 1 on the
magnitude scale. Thus, the amount of ground shaking for a magnitude 5 earthquake is 10 times greater
than that produced by a magnitude 4 earthquake.
In addition, each unit of Richter magnitude
equates to roughly a 32-fold energy increase. Thus, an
earthquake with a magnitude of 6.5 releases 32 times
as much energy as one with a magnitude of 5.5, and
roughly 1000 times as much energy as one with a
magnitude of 4.5 (Table 10.2). A major earthquake
with a magnitude of 8.5 releases millions of times
as much energy as the smallest earthquakes felt by
humans.
OTHER MAGNITUDE SCALES Richter’s original
goal was modest in that he only attempted to rank
the earthquakes of southern California (shallowfocus earthquakes) into groups of large, medium,
and small magnitude. Hence, Richter magnitude
was designed to study nearby (or local) earthquakes
and is denoted by the symbol (ML)—where M is for
magnitude and L is for local.
The convenience of describing the size of an
earthquake by a single number that could be calculated quickly from seismograms makes the Richter
scale a powerful tool. Further, unlike intensity scales
that could only be applied to populated areas of the
globe, Richter magnitudes could be assigned to earthquakes in more remote regions and even to events
that occurred in the ocean basins. As a result, the
10_tarb_ch10.indd 247
method devised by Richter was adapted to a number
of different seismographs located throughout the
world. In time, seismologists modified Richter’s work
and developed new magnitude scales. These modified scales were initially calibrated to be equivalent
to the Richter scale and have contributed to efforts
to measure the size of earthquakes. However, despite
their usefulness, none of these scales is adequate for
describing very large earthquakes.
In recent years seismologists have been employing a more precise measure
called moment magnitude (MW), which is derived
from the amount of displacement that occurs along
a fault zone rather than by measuring the ground
motion at some distant point. Moment magnitude
is calculated by using a combination of factors that
include the average amount of displacement along
the fault, the area of the rupture surface, and the
shear strength of the faulted rock—a measure of
how much strain energy a rock can store before it
suddenly slips and releases this energy in the form
of an earthquake (and heat). For example, the energy
involved in a 3-metre displacement of a rock body
along a rupture a few hundred kilometres long would
be much larger than that produced by a 1-metre
displacement along a 10-kilometre-long rupture
(assuming comparable rupture depths).
The moment magnitude can also be readily
calculated from seismograms by examining very
long period seismic waves. The values obtained have
been calibrated so that small- and moderate-sized
earthquakes have moment magnitudes that are
roughly equivalent to Richter magnitudes. However,
moment magnitudes are much better for describing
MOMENT MAGNITUDE
11/11/10 9:15 AM
248
Chapter 10
Earthquakes and Earth’s Interior
very large earthquakes. For example, on the moment
magnitude scale, the 1906 San Francisco earthquake,
which had a surface-wave magnitude of 8.3, would
be demoted to 7.9 on the moment magnitude scale,
whereas the 1964 Alaskan earthquake with an 8.3
Richter magnitude would be increased to 9.2. The
strongest earthquake on record is the 1960 Chilean
earthquake with a moment magnitude of 9.5.
Moment magnitude has gained wide acceptance among seismologists and engineers because
(1) it is the only magnitude scale that estimates
adequately the size of very large earthquakes; (2) it
is a measure established from the size of the rupture
surface and the amount of displacement; thus, it better reflects the total energy released during an earthquake; and (3) it can be verified by two independent
methods: field studies that are based on measurements of fault displacement and seismic methods by
using long-period waves.
EARTHQUAKE DESTRUCTION
Many factors determine the degree of destruction that will accompany an earthquake. The most
obvious is the magnitude of the earthquake and its
proximity to a populated area. Fortunately, most
earthquakes are small and occur in remote regions.
However, about 20 major earthquakes are reported
annually, one or two of which can be catastrophic.
Destruction from Seismic
Vibrations
The 1964 Alaskan earthquake provided geologists
with new insights into the role of ground shaking
as a destructive force. As the energy released by an
earthquake travels along Earth’s surface, it causes the
ground to vibrate in a complex manner by moving up
and down as well as from side to side. The amount
of structural damage attributable to the vibrations
depends on several factors, including (1) the intensity
of the vibrations, (2) the duration of the vibrations,
(3) the nature of the material on which the structure
rests, and (4) the design of the structure.
All of the multi-storey structures in Anchorage
were damaged by the vibrations. The more flexible wood-frame residential buildings fared best.
However, many homes were destroyed when the
ground failed. A striking example of how construction variations affect earthquake damage is shown in
Figure 10.14. You can see that the steel-frame building on the left withstood the vibrations, whereas
the poorly designed J. C. Penney building was
badly damaged. Engineers have learned that nonreinforced masonry buildings are the most serious
safety threat in earthquakes (see also Chapter opening photo and Figure 10.1A, B).
Most large structures in Anchorage were damaged, even though they were built according to the
earthquake building standards of the day. Perhaps
some of that destruction can be attributed to the
unusually long duration of this earthquake. Most
quakes consist of tremors that last less than a minute.
The Alaska quake was a notable exception: it reverberated for three to four minutes.
Although the
region within 20 to 50 kilometres of the epicentre
will experience about the same intensity of ground
shaking, the destruction varies considerably within
AMPLIFICATION OF SEISMIC WAVES
◆ Figure 10.14 Damage
caused to the five-storey
J. C. Penney Co. building,
Anchorage, Alaska. Very
little structural damage was
incurred by the adjacent
building.
(Courtesy of NOAA/Seattle)
10_tarb_ch10.indd 248
11/11/10 9:15 AM
Earthquake Destruction
this area. This difference is mainly attributable to
the nature of the ground on which the structures
are built. Soft sediments, for example, generally
amplify the vibrations more than solid bedrock.
Thus, the buildings located in Anchorage, which
were situated on unconsolidated sediments, experienced heavy structural damage. By contrast, most of
the nearby town of Whittier, although much closer
to the epicentre, rests on a firm foundation of granite and hence suffered much less damage. However,
Whittier was damaged by a tsunami (described in the
next section).
An earthquake that occurred in Mexico in 1985
gave seismologists and engineers a vivid reminder of
what had been learned from the 1964 Alaskan earthquake. The Mexican coast, where the earthquake was
centred, experienced unusually mild tremors despite
the strength of the quake. As expected, the seismic
waves became progressively weaker with increasing
distance from the epicentre. However, in the central
section of Mexico City, nearly 400 kilometres from
the source, the vibrations intensified to five times
that experienced in outlying districts. Much of this
amplified ground motion can be attributed to soft
sediments, remnants of an ancient lake bed, that
underlie portions of the city. The lake bed sediments
were shaken by the earthquake, causing buildings to
also shake and collapse (Figure 10.15).
249
◆ Figure 10.15 Effects of liquefaction. This tilted building
rests on unconsolidated sediment that behaved like
quicksand during the 1985 Mexican earthquake.
(Photo by James L. Beck)
In areas where unconsolidated
materials are saturated with water, earthquake vibrations can generate a phenomenon known as liquefaction (liqueo ⫽ to be fluid, facio ⫽ to make). Under
these conditions, what had been a stable soil turns
into a mobile fluid that is not capable of supporting buildings or other structures (Figure 10.15). As
a result, underground objects, such as storage tanks
and sewer lines, can literally float to the surface of
their newly liquefied environment. Buildings and
other structures can settle and collapse. During the
1989 Loma Prieta earthquake in California, foundations failed and geysers of sand and water shot from
the ground, indicating that liquefaction had occurred
(Figure 10.16).
LIQUEFACTION
Tsunami
Most deaths associated with the 1964 Alaskan quake
were caused by a tsunami* (tsu ⫽ harbour, nami ⫽
waves) or seismic sea wave. Before the catastrophic
tsunami of December 26, 2004 (Box 10.2), these
*Seismic sea waves were given the name tsunami by the Japanese, who
have suffered a great deal from them. The term tsunami is now used
worldwide.
10_tarb_ch10.indd 249
◆ Figure 10.16 These “mud volcanoes” were produced
by the Loma Prieta earthquake of 1989 in California. They
formed when fountains of sand and water shot from the
ground, an indication that liquefaction had occurred.
(Photo by Richard Hilton, courtesy of Dennis Fox)
11/11/10 9:15 AM
250
Chapter 10
BOX 10.2
Earthquakes and Earth’s Interior
U N D E R S T A N D IN G EA RTH
2004 Boxing Day Tsunami
Myanmar
Laos
India
Thailand
Region of
aftershocks
Sri
Lanka
m
tna
Vie
Burma plate
z on
ia
ys
ala
M
e
nch
m
at
ra
Epicentre of
December 26, 2004,
earthquake
Su
300 Kilometres
Cambodia
su bdu ct i o n
India plate
Sunda Tre
On December 26, 2004, the secondlargest earthquake ever recorded by
seismographs (9.3 MW) occurred 150
kilometres off the northwest coast
of Sumatra (Figure 10.D). The resulting tsunami damaged 11 countries
around the Indian Ocean and Andaman Sea, claimed more than 200,000
human lives, and left more than 1 million people homeless (Figure 10.E). At
the epicentre, located near the Sunda
Trench, the ocean floor was thrust up
to 5 metres and generated a wave
up to 50 metres high. This wall of
water smashed into the coasts of the
Seychelles Islands and Somalia, some
2000 kilometres away, and was even
recorded by tide gauges in California
and New Jersey. The south and east
coasts of Sri Lanka and India were
hit hard as their shorelines suddenly
◆ Figure 10.D This image highlights the crests (red) and troughs (blue) of the tsunami
generated by the great Sumatra earthquake of 2004. To the east of the Andaman and
Nicobar Islands, the waves travelled across the Andaman Sea to hit Thailand and Malaysia.
Simultaneously, the waves that were generated westward into the Indian Ocean hit Sri Lanka
(as depicted here),India, and the eastern coast of Africa.
(Modified from an image courtesy of the Geological Survery of Japan, AIST).
A
B
pulled back (Figure 10.F), exposing a large swath of sea bed as the
tsunami crested then surged its way
2 kilometres inland, destroying property and people.
The earthquake focus was approximately 10 kilometres beneath the
seafloor, generated by slippage deep
beyond the Sunda Trench where the
Australian-Indian plate subducts under
the Burma subplate, part of the Eurasian plate. The epicentre is located
near the intersection of two plates
(Eurasian and Australian-Indian) that
has generated several major earthquakes and tsunami, including the
1941 tsunami that killed 5000 people
along the east coast of India. Interactions between these plates were
also responsible for major volcanic
eruptions including Krakatoa, Tambora, and Toba. In spite of this record,
◆ Figure 10.E The village of Banda Aceh on
the north shore of Sumatra before (A) and after
(B) being hit by the 2004 tsunami.
(Photo by National Oceanic and Atmospheric
Administration)
10_tarb_ch10.indd 250
11/11/10 9:15 AM
Earthquake Destruction
251
A
B
◆ Figure 10.F
Aerial view of Kalutara
(south coast of Sri Lanka) before
(A) and during (B) shoreline withdrawal
just before the tsunami surged inland.
(Photos courtesy of Digital Globe/Getty
Images)
countries were taken by surprise when
the Boxing Day tsunami hit. This region
lacked the tide gauges, wave sensors,
regional tsunami warning system, and
communications network that the
Pacific region has (see Box 10.3). By
the time the Pacific network became
aware of the disaster, it was too late to
warn the Indian region, which underscored the need for a similar network.
destructive waves were commonly called tidal waves
by the media. However, that name is incorrect; the
waves are generated by earthquakes, not the tidal
effect of the Moon or Sun.
Most tsunami result from vertical displacement
along a fault located on the ocean floor (Figure 10.17),
or some other severe undersea disturbance, such as a
large volcanic eruption or submarine landslide. Once
created, a tsunami resembles the ripples formed when
a pebble is dropped into a pond. In contrast to ripples,
tsunami advance across the ocean at amazing speeds:
between 500 and 950 kilometres per hour. Despite this
striking characteristic, a tsunami in the open ocean
can pass undetected because its height is usually less
10_tarb_ch10.indd 251
In 2006 a DART tsunami buoy sensor
was deployed in the Indian Ocean to
collect wave data; by 2010 three buoys
and 17 seismic stations were installed
around the region.
than 1 metre and the distance between wave crests is
great, ranging from 100 to 700 kilometres. However,
on entering shallower coastal waters, these destructive
waves are slowed down and the water begins to pile up
to heights that occasionally exceed 30 metres (Figure
10.17). As the crest of a tsunami approaches the shore,
it appears as a rapid rise in sea level, with a turbulent
and chaotic surface. Tsunami can be very destructive
(see Boxes 10.2 and 10.3). Canadian examples include
the 1929 tsunami that killed 28 people and damaged
40 villages along the Burin Peninsula of southern
Newfoundland and Labrador (Figure 10.18A), and
1964 tsunami (generated by the great Alaskan earthquake) that caused extensive damage to Port Alberni
11/11/10 9:15 AM
252
Chapter 10
BOX 10.3
Earthquakes and Earth’s Interior
P E O P L E A N D TH E EN V IRON MEN T
Tsunami Warning System
Aleutian Is.
Asia
North
America
1h
2h
10
h 3h
8 h
4h
7 h
6 h
5h
5 h
4 h
Honolulu
h
3h
America
2h
1h
9
12
11 h
10 h
h
9
h
8
h
7
h
6
h
5h
4h
3
h
2
h
1
h
Tsunami traverse large stretches of
ocean before their energy is fully dissipated. The tsunami generated by
a 1960 Chilean earthquake, in addition to destroying villages along an
800-kilometre stretch of coastal South
America, travelled 17,000 kilometres
across the Pacific to Japan. Here, about
22 hours after the quake, considerable
damage occurred in south coastal villages. For several days afterward, tidal
gauges in Hilo, Hawaii, detected these
diminishing waves as they reverberated like echoes about the Pacific.
In 1946 a large tsunami struck the
Hawaiian Islands without warning.
A wave more than 15 metres high
left several coastal villages in shambles. This destruction motivated the
National Oceanic and Atmospheric
Administration to establish a tsunami
warning system for coastal areas of
the Pacific. From seismic observatories
throughout the region, large earthquakes are reported to the Pacific
Tsunami Warning Center at Ewa
Beach (near Honolulu), Hawaii. Scientists at the Center use tidal gauges
to determine whether a tsunami has
been formed. Within an hour a warning is issued. Although tsunami travel
very rapidly, there is sufficient time to
evacuate all but the region nearest the
New
Guinea
South
America
Australia
◆ Figure 10.G Tsunami travel times to Honolulu, Hawaii, from selected locations
throughout the Pacific.
(Data from NOAA)
epicentre (Figure 10.G). For example,
a tsunami generated near the Aleutian
Islands would take 5 hours to reach
Hawaii, and one generated near the
coast of Chile would travel 15 hours
before reaching Hawaii.
Tsunami speed:
835 km/h
Tsunami speed:
340 km/h
Fortunately, most earthquakes
do not generate tsunami. On the
average, only about 1.5 destructive
tsunami are generated worldwide
annually. Of these, only about 1 every
10 years is catastrophic.
Tsunami speed:
50 km/h
Sea level
Water depth:
5500 metres
Subducting
plate
Water depth:
900 metres
Water depth:
20 metres
Displacement
◆ Figure 10.17 Schematic drawing of a tsunami generated by displacement of the ocean floor. The speed of a wave correlates
with ocean depth. As shown, waves moving in deep water advance at speeds in excess of 800 kilometres per hour. Speed
gradually slows to 50 kilometres per hour at depths of 20 metres. Decreasing depth slows the movement of the wave. As waves
slow in shallow water, they grow in height until they topple and rush onto shore with tremendous force. The size and spacing of
these swells are not to scale.
10_tarb_ch10.indd 252
11/11/10 9:15 AM
Earthquake Destruction
A
253
◆ Figure 10.18 Examples of
destruction by tsunami in Canada.
A. Debris from houses, wharfs, boats,
and fishing gear along the shore of
the Burin Peninsula, Newfoundland
and Labrador, following the tsunami
generated by the Grand Banks
earthquake of 1929. B. Damage to
buildings associated with the tsunami
that hit Port Alberni, British Columbia,
in 1964.
(Photo A courtesy of the Rooms
Provincial Archives, A1–141/Parsons &
Sons; photo B courtesy of UBC Library,
Rare Books & Special Collections, neg #
AAC-15402)
B
on Vancouver Island (Figure 10.18B). The Burin
tsunami was triggered by a magnitude 7.2 earthquake
(epicentre 250 kilometres south of Newfoundland
and Labrador) that caused a turbidity current of 200
cubic kilometres to rush down the continental slope
of the Grand Banks while rupturing 12 transatlantic
cables and generating the tsunami. The wave rose to
27 metres and also reached the coasts of South
Carolina and Portugal.
Usually, the first warning of an approaching
tsunami is a relatively rapid withdrawal of water
from beaches (see Box 10.2). Coastal residents have
learned to heed this warning and move to higher
ground because about 5 to 30 minutes later, the
retreat of water is followed by a surge capable of
extending hundreds of metres inland. In a successive
fashion, each surge is followed by rapid oceanward
retreat of the water.
subsidence triggered by the vibrations. At Valdez
and Seward, the violent shaking caused deltaic
materials to experience liquefaction; the subsequent
slumping carried both waterfronts away. Because of
the threat of recurrence, the entire town of Valdez
was relocated about 7 kilometres away on more
stable ground. In Valdez, 31 people on a dock died
when it slid into the sea.
Most of the damage in the city of Anchorage
was also attributed to landslides. Many homes were
destroyed in Turnagain Heights when a layer of clay
lost its strength and more than 80 hectares of land
slid toward the ocean (Figure 10.19). A portion of this
spectacular landslide was left in its natural condition
as a reminder of this destructive event. The site was
appropriately named Earthquake Park. Downtown
Anchorage was also disrupted as sections of the main
business district dropped by as much as 3 metres.
Landslides and Ground Subsidence
Fire
In the 1964 Alaskan earthquake, the greatest damage to structures was from landslides and ground
The famous 1906 earthquake in San Francisco
reminds us of the formidable threat of fire. The
10_tarb_ch10.indd 253
11/11/10 9:15 AM
254
Chapter 10
Earthquakes and Earth’s Interior
Turnagain Heights
Layer of sand and gravel
Bootlegger Cove clay
Cracks developed
A
200 metres
Original profile
B
C
◆ Figure 10.19 Turnagain Heights slide caused by the 1964
Alaskan earthquake. A. Vibrations from the earthquake caused
cracks to appear near the edge of the bluff. B. Within seconds
blocks of land began to slide toward the sea on a weak layer of clay.
In less than five minutes, as much as 200 metres of the Turnagain
Heights bluff area had been destroyed. C. Photo of a small portion
of the Turnagain Heights slide.
(Photo C courtesy of U.S. Geological Survey)
central city contained mostly large older wooden
structures and brick buildings. Although many nonreinforced brick buildings were extensively damaged
by vibrations, the greatest destruction was caused by
fires when gas and electrical lines were severed. The
fires raged out of control for three days and devastated more than 500 blocks of the city. The problem
was compounded by the initial ground shaking,
which broke the city’s water lines into hundreds of
unconnected pieces.
The fire was finally contained when buildings
were dynamited along a wide boulevard to provide a
firebreak, the same strategy used in fighting a forest
fire. Although only a few deaths were attributed to
the San Francisco fire, such is not always the case. A
1923 earthquake in Japan triggered an estimated 250
fires, which devastated the city of Yokohama and
destroyed more than half the homes in Tokyo. More
than 100,000 deaths were attributed to the fires,
which were driven by unusually high winds.
CAN EARTHQUAKES BE
PREDICTED?
The earthquake that shook Northridge, California,
on January 17, 1994, was brief (about 40 seconds)
and only of moderate rating (MW 6.7), but it killed
10_tarb_ch10.indd 254
57 people and caused about US$40 billion in damage. The amount of devastation caused by the
earthquake was largely due to the fact that the area
it affected was densely populated and highly developed. Seismologists warn that earthquakes of comparable or greater strength will occur along the San
Andreas Fault system, which cuts a 1300-kilometre
path through the state. The obvious question is, can
earthquakes be predicted?
Short-Range Predictions
The goal of short-range earthquake prediction is
to provide a warning of the location and magnitude
of a large earthquake within a narrow time frame.
We currently do not have an absolutely reliable
method for making short-range earthquake predictions. However, substantial efforts to achieve this
objective are being put forth in Japan, the United
States, China, and Russia—countries where earthquake risks are high. This research has concentrated
on monitoring possible precursors—phenomena that
precede and thus provide a warning of a forthcoming
earthquake. In California, for example, seismologists
are measuring uplift, subsidence, and strain in the
rocks near active faults. Some Japanese scientists
are studying anomalous animal behaviour that may
precede a quake. Other researchers are monitoring
11/11/10 9:15 AM
Can Earthquakes Be Predicted?
changes in groundwater levels, while still others are
trying to predict earthquakes based on changes in the
electrical conductivity of rocks.
Among the most ambitious earthquake experiments is one being conducted along a segment of
the San Andreas Fault near the town of Parkfield in
central California. Here, earthquakes of moderate
intensity have occurred on a regular basis about once
every 22 years since 1857. The most recent rupture
was a magnitude 5.6 quake that occurred in 1966.
With the next event already significantly overdue,
the U.S. Geological Survey has established an elaborate monitoring network. Included are creepmeters,
tiltmeters, and borehole strain meters that are used
to measure the accumulation and release of strain.
Seventy seismographs of various designs have been
installed to record foreshocks and the main event.
Finally, a network of distance-measuring devices
that employ lasers measures movement across the
fault (Figure 10.20). The object is to identify ground
movements that may precede a sizeable rupture.
With this concept in mind, a group of seismologists plotted the distribution of rupture zones
associated with great earthquakes that have occurred
in the seismically active regions of the Pacific Basin.
The maps revealed that individual rupture zones
tended to occur adjacent to one another without
appreciable overlap, thereby tracing out a plate
boundary. Recall that most earthquakes are generated along plate boundaries by the relative motion of
large lithospheric blocks. Because plates are in continual motion, the researchers predicted that over
one or two centuries, major earthquakes would occur
along each segment of the Pacific plate boundary.
When the researchers studied historical
records, they discovered that some zones had not
produced a great earthquake in more than a century.
These quiet zones, called seismic gaps, were identified as probable sites for major earthquakes in the
next few decades (Figure 10.21).
1972
M = 7.4
Long-Range Forecasts
Fault
rupture,
1972
lt
fau
tte
arlo
Ch
een
Qu
In contrast to short-range predictions, which aim to
predict earthquakes within hours or at most days,
long-range forecasts give the probability of a certain
magnitude earthquake occurring on a time scale of
30 to 100 years or more. Long-range forecasts are
based on the premise that earthquakes are repetitive
or cyclic, like the weather. In other words, as soon
as one earthquake is over, the continuing motions of
Earth’s plates begin to build stress in the rocks again
until they fail once more. This has led seismologists
to study historical records of earthquakes, looking
for any discernible patterns so that the probability of
recurrence might be established.
255
ALASKA
Prince
Rupert
Fault rupture,
1949
Queen
Charlotte
Islands
BRITISH
COLUMBIA
Magnitude
Seismic
gap
8
7
Fault rupture,
1949 and 1970
1970
M=7
1929
M=7
6
5
4
◆ Figure 10.20 Lasers used to measure movement along
the San Andreas Fault.
(Photo by John K. Nakata/U.S. Geological Survey)
10_tarb_ch10.indd 255
◆ Figure 10.21 Map showing epicentres of large shallow
earthquakes recorded from 1929 to 1972 along the northwest
coast of British Columbia and the fault ruptures with which
they were associated. The seismic gap located immediately
west of the southern Queen Charlotte Islands denotes the
most likely location for the next large earthquakes along the
Queen Charlotte Fault.
(Courtesy of Tricouni Press Ltd.)
11/11/10 9:15 AM
256
Chapter 10
Earthquakes and Earth’s Interior
Another method of long-term forecasting,
known as paleoseismology (palaios ⫽ ancient, seismos ⫽
shake, ology ⫽ the study of), has been implemented.
One technique involves studying layered deposits
that were offset by prehistoric seismic disturbances.
Researchers recently discovered strong evidence that
very powerful earthquakes (magnitude of 8 or larger)
have repeatedly struck the Pacific Northwest over
the past several thousand years. The most recent
event occurred about 300 years ago. As a result
of these findings, public officials have taken steps
to strengthen some of the region’s existing dams,
bridges, and water systems.
In summary, it appears that the best prospects
for making useful earthquake predictions involve
forecasting magnitudes and locations on time scales
of years or perhaps even decades. These forecasts are
important because they provide information used to
develop earthquake building standards and to assist
in land-use planning.
PROBING EARTH’S INTERIOR
GEODe:Earth
Earth’s Interior
Earth’s Layered Structure
Earth’s interior lies just beneath us. However, direct
access to it remains very limited. Wells drilled
into the crust in search of oil, gas, and other
natural resources have generally been confined to the
upper 7 kilometres—only a small fraction of Earth’s
BOX 10.4
6370-kilometre radius. Even the Kola Well, a superdeep research well located in a remote northern outpost of Russia, has penetrated to a depth of only 12.3
kilometres. Although volcanic activity is considered
a window into Earth’s interior because materials are
brought up from below, it allows a glimpse of only
the outer 200 kilometres of our planet.
Fortunately, geologists have learned a great
deal about Earth’s composition and structure
through computer modelling, high-pressure laboratory experiments, and samples of the solar system
(meteorites) that collide with Earth. In addition,
many clues to the physical conditions inside our
planet have been acquired through the study of
seismic waves generated by earthquakes, nuclear
explosions, and weaker, human-produced vibrations. As seismic waves pass through Earth, they
carry information to the surface about the materials through which they were transmitted. Hence,
when carefully analyzed, seismic records provide an
image of materials below the surface. The Canadian
LITHOPROBE project, for example, has yielded
exciting insights on the deep lithosphere from the
processing of seismic data (see Box 10.4).
Much of our knowledge of Earth’s interior
comes from the study of earthquake waves that
penetrate Earth and emerge at some distant point.
The technique involves accurately measuring the
time required for P and S waves to travel from an
earthquake or nuclear explosion to a seismic station. Because the time required for P and S waves to
C A N A D IA N P R OFIL E
Lithoprobe: Probing the Depths of Canada
David Eaton*
Unlike the oceanic crust, continents
preserve a record of Earth’s early history, including how ancient mountains
formed and how the continents grew
and developed over time. Geoscientists are faced with a daunting challenge, however: to reconstruct this
tectonic record from fragmentary evidence provided by rocks exposed at
the surface. Canada’s LITHOPROBE
project, which commenced in 1984,
is a bold multi-institutional endeavour to unravel the tectonic evolution of North America with a focus
on the critical third dimension of
10_tarb_ch10.indd 256
geology: depth. With collaboration
by more than 900 scientists working
in academia, government agencies,
and the petroleum and mining industries, LITHOPROBE has involved a
significant segment of Canada’s Earth
science community in a coordinated
research effort. The research program
was divided into a series of transects
that collectively span the evolution of
the North American continent in time
(about 4 billion years) and space.
To achieve adequate resolution
of geologic features down to the
base of the crust (approximately 40
kilometres) and the upper mantle,
LITHOPROBE research has been
spearheaded by multichannel seismicreflection (MSR) profiling. Adapted
from the petroleum industry, where it
is used extensively for exploration of
the top 10 kilometres of sedimentary
basins, this technique makes use of
P waves, usually generated mechanically by using vibroseis trucks, to
record faint echoes produced at
buried geologic contacts, faults, and
shear zones. Much like medical imaging techniques, the seismic data are
processed by computer to enhance
the seismic reflections, which produces cross-sectional images that can
11/11/10 9:15 AM
Probing Earth’s Interior
Cadwallader
Bridge R.
Belt of volcanoes
Terrane
Terrane
Pacific Rim
Shuksan(e.g., Garibaldi)
Harrison
Terrane
Methow
Nanaimo
Terrane
terranes
sediments
Crescent
Ocean Deformation Tofino Basin Terrane
sediments
front
sediments
0
20
Oceanic crust
Lower lithosphere
(upper mantle)
Depth (km)
40
Accrete
dw
Juan edge
de
Fuc
a
60
Oceanic asthenosphere
(a few percent molten rock)
80
100
120
granite
plutons
Wrangellia
Pre-T
ertiar
y und
erplat
Tert
e
iary
und
Co
e
rpla
ld o
te
cea
nic
lith
osp
her
e
(old oceanic crust, island
arcs and sediments accreted
during Cretaceous Period)
pla
257
te
Partial melting of
Juan de Fuca plate
140
mafic residue
Trapped remnants Subcrustal
of older,
lithosphere
underthrusted
oceanic plate
Intermontane
Composite
Terrane
(Lower Crust)
Continental
asthenosphere
(a few percent
molten rock)
160
Pacific
Ocean
Offshore
Slope/Shelf
SW
Insular
Belt
Vancouver Island
Coast
Belt
Strait of
Georgia
NE
◆ Figure 10.H Part of the Canadian LITHOPROBE project involved the study of the lithospheric structure of western Canada. Among the
exciting discoveries was the detection of lithospheric fragments (underplates) that remain lodged under the west coast region and have
contributed to the growth of mountain roots.
(Modified from image courtesy of R. M. Clowes)
be compared with geologic data to
reveal the underlying anatomy of the
lithosphere. Specialists in all areas of
geology, geophysics, and geochemistry have collaborated on major
geotectonic problems by using techniques that include basic mapping,
precise geochronology, electromagnetic studies, seismic refraction, and
paleomagnetism.
LITHOPROBE’s pilot study on Vancouver Island demonstrated that MSR
profiling methods can be used to map
layers near the top of the descending Juan de Fuca slab (Figure 10.H).
It showed slabs of lithosphere welded
to the continental lithosphere of west-
ern British Columbia as underplates.
Results from this modern subduction
system (Cascadia) have provided a
template for understanding ancient
subduction systems. Furthermore, subduction “scars” have been detected in
the mantle beneath Archean rocks of
the Superior craton in Quebec and
the Proterozoic Great Bear Arc in the
Northwest Territories. These results
establish that the subduction occurred
in some areas in the late Archean and
support the concept that subducted
slabs can be accreted to the base of
continental plates as part of the process of formation of deep lithospheric
roots.
travel through Earth depends on the properties of
the materials encountered, seismologists search for
variations in travel times that cannot be accounted
for simply by differences in the distances travelled.
These variations correspond to changes in the properties of the materials encountered.
One major problem is that to obtain accurate
travel times, seismologists must establish the exact
location and time of an earthquake. This is commonly a difficult task because most earthquakes occur
10_tarb_ch10.indd 257
LITHOPROBE’s main research
focused on ancient shield and platform
regions where North America is known
to have grown via a series of mountainbuilding events. LITHOPROBE profiles
have allowed geoscientists to trace
structural contacts down to the base
of the crust (Moho). This work has led
to the discovery of previously unrecognized mountain belts that lie buried beneath undeformed sedimentary
rocks.
For further information visit
www.lithoprobe.ca.
*David Eaton is a professor in the Department
of Earth Sciences at the University of Calgary.
in remote areas. By contrast, the time and location
of a nuclear test explosion is always known precisely.
Despite the limitations of studying seismic waves
generated by earthquakes, seismologists in the first
half of the twentieth century were able to use them to
determine the major layers of Earth. It was not until
the early 1960s, when nuclear testing was in its heyday and networks consisting of hundreds of sensitive
seismographs were deployed, that the finer structures
of Earth’s interior were established with certainty.
11/11/10 9:15 AM
258
Chapter 10
Earthquakes and Earth’s Interior
The Nature of Seismic Waves
Earthquake source
To examine Earth’s composition and structure, we
must first study some of the basic properties of wave
transmission, or propagation. As stated earlier in this
chapter, seismic energy travels out from its source
in all directions as waves. For purposes of description, the common practice is to consider the paths
taken by these waves as rays, or lines drawn perpendicular to the wave front, as shown in Figure 10.22.
Significant characteristics of seismic waves include
the following:
• The velocity of seismic waves depends on the
density and elasticity of the intervening material.
Seismic waves travel most rapidly in rigid materials that elastically spring back to their original
shapes when the stress caused by a seismic wave is
removed. For instance, crystalline rock transmits
seismic waves more rapidly than does a layer of
unconsolidated mud.
• Within a given layer, the speed of seismic waves
generally increases with depth because pressure
increases and squeezes the rock into a more compact elastic material.
• Compressional waves (P waves), which vibrate
back and forth in the same line as their direction
of travel, are able to propagate through liquids
as well as solids because when compressed, these
materials behave elastically; that is, they resist a
change in volume and, like a rubber band, return
to their original shape as a wave passes.
• Shear waves (S waves), which vibrate at right
angles to their direction of travel, cannot propagate through liquids because, unlike solids, liquids
have no shear strength. That is, when liquids are
subjected to forces that act to change their shapes,
they simply flow.
• In all materials, P waves travel faster than S waves.
Wave
fronts
Rays
◆ Figure 10.22 Seismic energy travels in all directions from
an earthquake source (focus). The energy can be portrayed as
expanding wave fronts or as rays drawn perpendicular to the
wave fronts.
These measurable changes in seismic wave motions
enable seismologists to probe Earth’s interior.
Seismic Waves and Earth’s
Structure
If Earth were a perfectly homogeneous body, seismic waves would spread through it in all directions.
Such seismic waves would travel in a straight line at
a constant speed. However, this is not the case for
Earth. It so happens that the seismic waves reaching
seismographs located farther from an earthquake
travel at faster average speeds than those recorded at
locations closer to the event. This general increase in
speed with depth is a consequence of increased pressure, which enhances the elastic properties of deeply
• When seismic waves pass from one material to
another, the path of the wave is refracted (bent).*
In addition, some of the energy is reflected from
the discontinuity (the boundary between the two
dissimilar materials). This is similar to what happens to light when it passes from air into water.
Thus, depending on the nature of the layers
through which they pass, seismic waves speed up or
slow down and can be refracted (bent) or reflected.
*Refraction occurs provided that the rnay is not travelling perpendicular
to the boundary.
10_tarb_ch10.indd 258
◆ Figure 10.23 Wave paths through a planet where
velocity increases with depth.
11/11/10 9:15 AM
Discovering Earth’s Major Boundaries
Inner
core
Outer
core
Mantle
◆ Figure 10.24 A few of the many possible paths that
seismic rays take through Earth.
buried rock. As a result, the paths of seismic rays
through Earth are refracted in the manner shown in
Figure 10.23.
As more sensitive seismographs were developed, it became apparent that in addition to gradual
changes in seismic-wave velocities, rather abrupt
velocity changes also occur at particular depths.
Because these discontinuities were detected worldwide, seismologists concluded that Earth must be
composed of distinct shells that have varying compositions or mechanical properties (Figure 10.24; see
also Figure 1.19).
DISCOVERING EARTH’S
MAJOR BOUNDARIES
Over the past century, seismic data gathered from
many seismic stations have been compiled and
analyzed. From this information, seismologists
have developed a detailed image of Earth’s interior (Figure 1.19). This model is continually being
fine-tuned as more data become available and as
new seismic techniques are employed. Furthermore,
laboratory studies that experimentally determine
the properties of various Earth materials under the
extreme environments found deep in Earth add to
this body of knowledge.
The Crust–Mantle Boundary
(the Moho)
In 1909 a pioneering Yugoslavian seismologist,
Andrija Mohorovičić, presented the first convincing
evidence for layering within Earth. The boundary
10_tarb_ch10.indd 259
259
he discovered separates crustal materials from rocks
of different composition in the underlying mantle,
and it was named the Mohorovičić discontinuity in
his honour. The name for this boundary was quickly
shortened to Moho.
By carefully examining the seismograms of
shallow earthquakes, Mohorovičić found that seismic
stations located more than 200 kilometres from an
earthquake obtained appreciably faster average travel
velocities for P waves than did stations located nearer
the quake (Figure 10.25). In particular, P waves that
reached the closest stations first had velocities that
averaged about 6 kilometres per second. By contrast,
the seismic energy recorded at more distant stations
travelled at speeds that approached 8 kilometres per
second. This abrupt jump in velocity did not fit the
general pattern that had been previously observed.
From these data, Mohorovičić concluded that below
50 kilometres, a layer exists with properties markedly
different from those of Earth’s outer shell.
Figure 10.25 illustrates how Mohorovičić
reached this important conclusion. Notice that
the first wave to reach the seismograph located
100 kilometres from the epicentre travelled the
shortest route directly through the crust. However,
at the seismograph located 300 kilometres from the
epicentre, the first P wave to arrive travelled through
the mantle, a zone of higher velocity. Thus, although
this wave travelled a greater distance, it reached the
recording instrument sooner than did the rays taking
the more direct route. This is because a large portion
of its journey was through a region having a composition where seismic waves travel more rapidly. This
principle is analogous to a driver taking a bypass
route around a large city during rush hour. Although
this alternative route is longer, it may be faster.
The Core–Mantle Boundary
A few years later, in 1914, the location of another
major boundary was established by the German
seismologist Beno Gutenberg.* This discovery was
based primarily on the observation that P waves
diminish and eventually die out completely about
105° from an earthquake (Figure 10.26). Then,
about 140° away, the P waves reappear, but about
two minutes later than would be expected based on
the distance travelled. This belt, where direct seismic
waves are absent, is about 35° wide and has been
named the P-wave shadow zone** (Figure 10.26).
*The core–mantle boundary had been predicted by R. D. Oldham
in 1906, but his arguments for a central core did not receive wide
acceptance.
**As more sensitive instruments were developed, weak and delayed P
waves that enter this zone via reflection were detected.
11/11/10 9:15 AM
260
Chapter 10
Earthquakes and Earth’s Interior
Seismic station 1
100 km
Focus
w
hallo
Seismic station 2
200 km
Seismic station 3
300 km
Crust
(average P-wave
velocity is 6 km/s)
es
wav
S
Moho
Upper mantle
(average P-wave
velocity is 8 km/s)
Deep waves
A Time 1—Slower shallow waves arrive at seismic station 1 first.
Seismic station 2
200 km
Seismic station 1
100 km
Focus
Shallow
waves
Seismic station 3
300 km
Crust
t
Moho
Deep waves
Mantle
B Time 2—Slower shallow waves arrive at seismic station 2 first.
Seismic station 1
100 km
Seismic station 2
200 km
Shallow waves
Seismic station 3
300 km
Crust
Focus
Moho
Deep waves
Mantle
C Time 3—Faster deeper waves arrive at seismic station 3 first.
◆ Figure 10.25 Idealized paths of seismic waves travelling from an earthquake focus to three seismic stations. In parts A and B,
you can see that the two nearest recording stations receive the slower waves first because the waves travelled a shorter distance.
However, as shown in part C, beyond 200 kilometres, the first waves received passed through the mantle, which is a zone of
higher velocity.
Gutenberg, and others before him, realized
that the P-wave shadow zone could be explained if
Earth contained a core that was composed of material unlike the overlying mantle. The core, which
Gutenberg calculated to be located at a depth of
2900 kilometres, must somehow hinder the transmission of P waves, similar to the way in which light
rays are blocked by an object that casts a shadow.
However, rather than actually stopping the P waves,
10_tarb_ch10.indd 260
the shadow zone is produced by bending (refracting)
of the P waves, which enter the core, as shown in
Figure 10.26.
It was further determined that S waves do not
travel through the core. This fact led geologists to
conclude that at least a portion of this region is liquid. This conclusion was further supported by the
observation that P-wave velocities suddenly decrease
by about 40 percent as they enter the core. Because
11/11/10 9:15 AM
Chapter Summary
261
Earthquake epicentre
Key
P wave
S wave
P wave
S wave
Seismic station records
both P and S waves
105°
sh P-w
ad av
ow e
zo
ne
ve e
wa on
P- ow z
ad
sh
105°
Seismic station records
no P or S waves
P wave
140°
140°
S-wav
Seismic station records
P waves only
ne
e shado w zo
◆ Figure 10.26 View of Earth’s interior showing P-wave and S-wave paths. Any location more than 105° from the earthquake
epicentre will not receive direct S waves because the outer core will not transmit them. Although P waves are also absent beyond
105°, they are recorded beyond 140° because of refraction.
melting reduces the elasticity of rock, this evidence
pointed to the existence of a liquid layer below the
rocky mantle.
Discovery of the Inner Core
In 1936 the last major subdivision of Earth’s interior
was predicted by Inge Lehmann, a Danish seismologist. Lehmann discovered a new region of seismic
reflection and refraction within the core. Hence, a
core within a core was discovered. The size of the
inner core was not precisely established until the
early 1960s, when underground nuclear tests were
conducted in Nevada. Because the exact location and
time of the explosions were known, echoes from seismic waves that bounced off the inner core provided a
precise means of determining its size.
From these data, the inner core was found to
have a radius of about 1216 kilometres. Furthermore,
P waves passing through the inner core have appreciably faster average velocities than do those penetrating only the outer core. The apparent increase
in the elasticity of the inner core is evidence that this
innermost region is solid.
CHAPTER SUMMARY
• Earthquakes occur when crustal rocks rupture
at the focus, below the epicentre. Rupture generates seismic waves and occurs by elastic rebound
to release built-up stress along a fault. Foreshocks
often precede a major earthquake and aftershocks
follow it.
10_tarb_ch10.indd 261
• Seismic waves include body (through the Earth)
and surface (around the Earth). Body waves include
faster primary or P waves that travel through rock
along a direct path and secondary or S waves that
shake rock at right angles to travel direction.
11/11/10 9:15 AM
262
Chapter 10
Earthquakes and Earth’s Interior
• Epicentres are located using travel times of P and
S waves and triangulating distances from three
or more seismic stations. Most epicentres occur
along the circum-Pacific belt and the oceanic ridge
system.
• The Modified Mercalli Intensity Scale is based on
damage to buildings. Magnitude estimates energy
released using the Richter scale by comparing largest wave amplitude with distance to the epicentre. Each higher number means 32 times more
energy was released.
• Structural damage depends on wave amplitude,
duration of vibrations, substrate (geology), and
structural design. Secondary effects of quakes
include tsunami, landslides, subsidence, and fire.
• While short-range predictions are not yet possible,
earthquake-prone regions with seismic gaps (overdue) are well known. P and S waves travel faster
in solid elastic materials, slower in weaker layers,
and are reflected and refracted (bent) at material
boundaries. This tells us that Earth’s layers are
crust, mantle, outer liquid, and inner solid core.
Visit www.pearsoned.ca/mygeoscienceplace to find chapter review quizzes,
GEODe: Earth activities, Interactive Animations, and much more.
REVIEW QUESTIONS
1. What is an earthquake? Under what circumstances do earthquakes occur?
2. How are faults, foci, and epicentres related?
3. Explain what is meant by elastic rebound.
4. Describe the principle of a seismograph.
5. List the major differences between P (primary),
S (secondary), and L (surface or long) waves.
Which of these seismic wave types causes the
greatest destruction?
6. Each increase of 1 on the Richter scale corresponds to a _____-fold increase in wave amplitude and a _____-fold increase in energy release.
10_tarb_ch10.indd 262
7. In addition to seismic vibrations, what other factors contribute to the destruction associated with
earthquakes?
8. What is a tsunami? How is one generated?
9. Explain how the following boundaries within
Earth’s interior were discovered:
a) The crust-mantle boundary (Mohorovičić
discontinuity or Moho)
b) The core-mantle boundary
c) The inner core–outer core boundary
10. What types of information are useful in making
short-range predictions versus long-range forecasts of future earthquake activity?
11/11/10 9:16 AM